강의

멘토링

커뮤니티

AI Technology

/

Deep Learning & Machine Learning

Triangles in action! OpenAI Triton beginner

In this course, you will learn how to program kernels and develop PyTorch modules. You can use the knowledge you have gained to develop models faster.

(4.1) 7 reviews

77 learners

Level Basic

Course period Unlimited

  • triangle
Deep Learning(DL)
Deep Learning(DL)
Python
Python
gpgpu
gpgpu
triton
triton
openai
openai
Deep Learning(DL)
Deep Learning(DL)
Python
Python
gpgpu
gpgpu
triton
triton
openai
openai

Reviews from Early Learners

Reviews from Early Learners

4.1

5.0

손동휘

100% enrolled

It teaches very easily but effectively~!!

5.0

book1220

19% enrolled

There were a lot of parts I didn't understand, but listening to Samgak-nim's lecture helped me a lot in understanding the basic concepts. ㅎㅎ

5.0

ㅇㅅㅇ

100% enrolled

You teach Triton fundamentals very comprehensively. After explaining & practicing Triton, you provide & explain various examples based on CNN. By practicing based on this, you can gain deep understanding. The examples in the latter part require experience training CNN with PyTorch to understand. Although CNN is explained, it feels more like a review. However, even if you're not familiar with PyTorch, you can sufficiently build Triton fundamentals through just the first part of the lectures!

What you will gain after the course

  • OpenAI Triton Programming

  • Derivation of backpropagation formula

  • PyTorch automatic differentiation engine

  • Implementing new operations in PyTorch

First release worldwide!
GPU Expert Triangle Explains
OpenAI Triton, complete conquest!

If you are a deep learning developer
This is the OpenAI Triton you absolutely must know.

Hello! This is Triangle.
Are you spending too much time training and inferring deep learning models?
Well then, it's time to write your own kernel .
Write kernels with OpenAI Triton and accelerate model training and inference times!


Experience the power of OpenAI Triton!

Implementing Softmax with OpenAI Triton can achieve faster performance than PyTorch.
Check out the performance of PyTorch, PyTorch JIT, and OpenAI Triton in the table below.
You can see that OpenAI Triton has the strongest performance!


This course is for beginners!

No prior knowledge of GPU programming is required.
However, you must know how to use Python and PyTorch .


I recommend this course to these people!

  1. Anyone who is interested in GPU programming and wants to learn
  2. Deep Learning Model Developer
  3. Anyone who wants to experience OpenAI Triton

If you take the course, you will learn about OpenAI Triton
You will be reborn as a person of ability.

  • You can get basic knowledge about OpenAI Triton.
  • You can write any operation you want in OpenAI Triton.
  • We can derive the backpropagation formula.
  • Understand PyTorch's automatic differentiation engine.
  • You can implement operations that are not available in PyTorch.

Easy explanations and high-quality lectures
Provides.

We provide easy-to-understand explanations and high-quality lectures to help you quickly learn OpenAI Triton.
I write the lecture materials and code myself .

자동 미분 엔진


If you are a student or job seeker ,
Get a student discount!

Request a student discount via email, Discord, or inquire before enrolling.
We will ease your financial burden!

✅ Information before taking the course

  • Examples for the lecture are shared via GitHub (link) .
  • If you have any questions about anything not covered in the lecture, you can ask them through Discord (link) . We also provide a channel where you can discuss OpenAI Triton with other students.

Recommended for
these people

Who is this course right for?

  • Anyone interested in GPU programming and want to learn

  • Deep Learning Model Developer

  • Anyone interested in experiencing OpenAI Triton

Need to know before starting?

  • Python

  • PyTorch

Hello
This is

366

Learners

39

Reviews

47

Answers

4.3

Rating

3

Courses

I have been working in the software field for over 10 years. I am deeply interested in graphics, parallel computing, and artificial intelligence, and I strive to develop my expertise in these areas. Having worked at Samsung Electronics, AMD, LINE Plus, and Kakao Brain, I am currently at Qualcomm. Although I did not join due to personal reasons, I have also successfully passed final interviews at Imagination Technologies, 42dot, Hyperconnect, NAVER LABS, Amazon, Google, and NVIDIA. Moving forward, I would like to share the knowledge I have gained with you and learn much from you as well. 🙇

Curriculum

All

16 lectures ∙ (3hr 5min)

Course Materials:

Lecture resources
Published: 
Last updated: 

Reviews

All

7 reviews

4.1

7 reviews

  • donghwee04025239님의 프로필 이미지
    donghwee04025239

    Reviews 1

    Average Rating 5.0

    5

    100% enrolled

    It teaches very easily but effectively~!!

    • dnjstlr12204011님의 프로필 이미지
      dnjstlr12204011

      Reviews 2

      Average Rating 5.0

      5

      19% enrolled

      There were a lot of parts I didn't understand, but listening to Samgak-nim's lecture helped me a lot in understanding the basic concepts. ㅎㅎ

      • triangle
        Instructor

        Glad you found it helpful! I think Triton can be a powerful weapon for developing deep learning models.

    • nemo님의 프로필 이미지
      nemo

      Reviews 4

      Average Rating 5.0

      Edited

      5

      100% enrolled

      You teach Triton fundamentals very comprehensively. After explaining & practicing Triton, you provide & explain various examples based on CNN. By practicing based on this, you can gain deep understanding. The examples in the latter part require experience training CNN with PyTorch to understand. Although CNN is explained, it feels more like a review. However, even if you're not familiar with PyTorch, you can sufficiently build Triton fundamentals through just the first part of the lectures!

      • gusdnd8522427님의 프로필 이미지
        gusdnd8522427

        Reviews 1

        Average Rating 5.0

        Edited

        5

        81% enrolled

        This lecture was extremely helpful in understanding Triton and the basics of GPU programming. Pros: 1. Overall, the lecture videos are not too long and the explanations are easy to understand, which was great for someone starting out lightly. When you're trying a new field for the first time, if the lecture videos are several hours long and the explanations are too difficult, you get tired quickly. 2. The lecture materials include appropriate visualizations that were very helpful in understanding how GPUs and various functions work. 3. Theory and practice are appropriately mixed, so you can both learn the theory and understand how that theory is actually implemented in code. There aren't really any major downsides, but if I had to pick something, the practice code isn't explained in enough detail for a complete beginner to easily understand just from watching once. Of course, you do provide general explanations. But these days ChatGPT explains things so well that I could immediately ask questions about things I was curious about, like why certain lines in the code had to be written that way, and study while getting answers right away, so this part wasn't really inconvenient. I really enjoyed the lecture and thank you for creating such a good course! 😊

        • skim0863님의 프로필 이미지
          skim0863

          Reviews 1

          Average Rating 5.0

          5

          31% enrolled

          $115.50

          triangle's other courses

          Check out other courses by the instructor!

          Similar courses

          Explore other courses in the same field!